EU AI Act and Global Regulation: What Deployers Must Track

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-ethics-safety-eu-ai-act-global-regulation-adults

1. A company based in Canada develops an AI-powered hiring tool that it sells exclusively to companies in Brazil and Australia. Under what circumstances would this company become subject to the EU AI Act?

   1. If the company advertises in EU member states
   2. If any of the Brazilian or Australian clients have employees working in the EU
   3. If the company uses EU-based servers to process data
   4. If the company plans to expand to the EU within five years

2. Which of the following AI systems would be classified as 'limited risk' under the EU AI Act?

   1. A chatbot integrated into a customer service website
   2. An AI system that scores credit applications for a European bank
   3. A real-time facial recognition system used for public security
   4. An AI-powered medical diagnosis tool used in hospitals

3. According to the regulatory timeline, when do the prohibitions on unacceptable risk AI practices take effect?

   1. August 2027
   2. August 2026
   3. February 2025
   4. August 2025

4. Under the EU AI Act, which entity bears primary responsibility for compliance with GPAI (general purpose AI) model requirements such as adversarial testing and incident reporting?

   1. The model provider who develops and releases the foundation model
   2. The end-user who interacts with the AI
   3. The deployer who builds applications using the model
   4. The cloud infrastructure provider hosting the model

5. Which of the following is explicitly banned under the EU AI Act's unacceptable risk category?

   1. AI-powered spam filters in email clients
   2. AI-generated content detection tools
   3. Recommendation algorithms on streaming platforms
   4. Social scoring systems used by government authorities

6. What specific obligations does the EU AI Act impose on providers of very large foundation models that are deemed to have 'systemic risk'?

   1. They must make their model weights fully open source
   2. They must conduct adversarial testing, report incidents to EU authorities, disclose energy consumption, and implement cybersecurity measures
   3. They must limit model capabilities to below a certain performance threshold
   4. They must submit to annual audits by EU military agencies

7. A fintech startup in Germany wants to deploy an AI system that analyzes transaction patterns to detect fraud. What risk tier and key requirements would apply?

   1. Unacceptable risk—the system cannot be deployed
   2. High-risk—requires conformity assessment, human oversight, data governance, and post-market monitoring
   3. Limited risk—transparency disclosures required
   4. Minimal risk—no mandatory requirements apply

8. How does the current US federal approach to AI regulation compare to the EU AI Act?

   1. The US requires all AI systems to undergo mandatory conformity assessments before deployment
   2. The US has banned all AI systems that the EU classifies as unacceptable risk
   3. The US has enacted a comprehensive federal AI law covering all risk tiers
   4. The US relies on sector-specific guidance from agencies like FTC, EEOC, and CFPB, plus an executive order

9. What approach has the United Kingdom taken to AI regulation?

   1. A complete prohibition on generative AI systems
   2. A mandatory licensing system for all AI developers
   3. A sector-based approach where existing regulators apply their domain rules to AI
   4. A comprehensive national AI act modeled on the EU AI Act

10. China has implemented specific regulations for AI systems. Which statement accurately reflects China's regulatory approach?

    1. China prohibits all forms of algorithmic recommendation systems
    2. China has no regulations on AI whatsoever
    3. China requires mandatory security assessments for generative AI models released to the public
    4. China has adopted the EU AI Act as its primary AI regulation

11. A company is developing an AI-powered resume screening tool for hiring. Under the EU AI Act, what would be the primary compliance requirement before deploying this system in the EU market?

    1. The company must conduct a conformity assessment demonstrating compliance with high-risk requirements
    2. The system must be approved by the European Commission before any deployment
    3. The company must open-source the algorithm to the public
    4. The system must be certified as safe by an EU member state military agency

12. Which of the following best describes the current state of AI regulation in Brazil?

    1. Brazil has fully implemented a comprehensive AI law
    2. Brazil defers entirely to EU AI Act for all AI governance
    3. Brazil has no AI regulations and no pending legislation
    4. Brazil has an AI bill pending that is modeled partly on GDPR and the EU AI Act

13. An AI-powered video game uses machine learning for non-player character behavior. How would this be classified under the EU AI Act?

    1. Limited risk, requiring transparency disclosures
    2. Unacceptable risk, due to simulated intelligence
    3. Minimal risk, as it falls under AI in video games
    4. High-risk, due to potential psychological effects on users

14. What is the deadline for full implementation of the EU AI Act across all provisions?

    1. February 2025
    2. August 2025
    3. August 2027
    4. August 2026

15. Which of the following is a transparency obligation under the EU AI Act for limited risk systems?

    1. Providing detailed technical documentation to regulators upon request
    2. Implementing human oversight controls for every decision
    3. Disclosing to users that they are interacting with an AI system and informing them about synthetic content generation
    4. Eliminating all bias from the AI system